The size and frequency of meals eaten by mammals appear to be controlled at least partly by neural and endocrine signals originating from the gut and other visceral organs. Serotonin (5-hydroxytryptamine; 5-HT) is an indoleamine that is synthesized by neurons in the peripheral and central nervous systems and by endocrine cells in the gastrointestinal mucosa. The anatomy and physiology of peripheral 5-HT suggest that this amine may serve a neuroendocrine role as a short-term control of feeding. Consistent with that hypothesis, recent pharmacological studies revealed that parenterally administered 5-HT inhibits feeding induced by food deprivation. In this laboratory, subcutaneous (SC) or intraperitoneal (IP) injection of 5-HT produced dose-related decreases in food intake. 5-HT inhibited feeding more potently after IP than than after SC injection in rats. Given IP, the dose of 5-HT that reduced feeding by 50 percent (RD-50) was behaviorally selective because it did not: inhibit drinking induced by water deprivation; elicit drinking in sated rats; impair exploration in an an open field; or act as an unconditional stimulus for the formation of a conditioned taste aversion. Total subdiaphragmatic vagotomy failed to alter the RD-50, slope or maximal anorexic effect of 5-HT. The anorexic action of 5-HT was antagonized by methysergide and by ketanserin. Taken together, the data suggest that gut 5-HT serves a paracrine or hormonal role as an extravagal signal for preabsorptive satiety. Furthermore this signal appears to be mediated at least partly by a ketanserin-sensitive receptor mechanism. The proposed investigation attempts to systematically evaluate the mechanisms by which IP administration of 5-HT inhibits feeding by determining: 1) whether 5-HT elicits behaviors characteristic of normal satiety in intact rats and in gastric shamfeeding rats; 2) the extent to which 5-HT inhibits feeding without altering other appetitive or consummatory responses; 3) whether denervating the liver, ablating the area postrema or removing neurosensory input from the abdominal viscera to the brain prevents the anorexic action of peripherally administered 5-HT; 4) whether infusing 5-HT intraportally inhibits feeding in rats; and 5) the pharmacological mechanisms for the anorexic action of 5-HT. In concert, these studies should provide significant information regarding the roles of peripheral 5-HT in normal and pathological processes related to the controls of feeding and gut function.